Digit Recognition of digital meters using TensorFlow 2 Object Detection API
This repository showcases the digit recognition on the utility meters with TensorFlow2 Object Detection API using the MobileNet V1 model.
This repository addresses the real challenge of tracking and monitoring the meter reading in the utility and asset-intensive manufacturing or refining industry.
The trained workforce is deployed to collect the meter readings from the site and then collect the reading manually in spreadsheets. This traditional practice is prone to human errors and sometimes it is very inefficient and costly.
This repository is covering the business case of detecting the digits from the images of digital meters. We can use this model to read the digits automatically from the images and store the result in the database. I am only covering the detection part in my repository.
For more details, please refer to the associated article - https://medium.com/@gauravsaini.728/digit-recognition-of-digital-meters-using-tensorflow-2-object-detection-api-48364cd678a9